Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Well, that's why i use Protocol A so i have asynchronous sync. I tested using FIO, too. All my results there around 40mb/s with drbd disconnected. Once I started DRBD I had round about 500kb/s in all of my tests. So the problem isn't the throughput of the network. That's why I'm asking, if there is any known issue or workaround regarding this. Kind regards, Tim Rohwedder -----Ursprüngliche Nachricht----- Von: Sebastian Riemer [mailto:sebastian.riemer at profitbricks.com] Gesendet: Donnerstag, 25. April 2013 15:43 An: GG-Net Mailing Cc: drbd-user at lists.linbit.com Betreff: Re: [DRBD-user] Performance Issues on version 8.3.13 On 25.04.2013 14:04, GG-Net Mailing wrote: > Hello, > > i'm running drbd on two Debian Wheezy x64 servers. > > I'm using protocol A for asynchronous sync, since one of the machines has a hybrid RAID10 on an adaptec 71605Q. Build of 2x 3TB Seagate Constellation CS SED and 2x Samsung 840 500GB. And the other only has a plain 1TB HDD. > > If I run a simple dd test on the node with the RAID copying from /dev/zero to a file on the device with drbd disabled, I get results around 400mb/s. > > The moment I start drbd with the node as primary and run the test again, my results are roughly 110mb/s. > > Are there any known issues regarding this? Yes, don't expect wonders from your 1 GbE network link or the single HDD on the secondary! 1 Gbit/s / 8 = 128 MB/s. More isn't possible with 1 GbE. If disconnected, the data doesn't have to travel through the network to the secondary. Simple as that. Cheers, Sebastian